The Problem – GIGO
NOTE: The purpose of the next few lessons is to help you use AI to get better results from your thinking.
If you need to open your own personal ChatGPT account there is no fee. ACTION: Click here https://openai.com/product and go to MENU then LOGIN.
You may also use other AIs like Copilot, Claude, Perplexity etc.
Let’s get started. The written conversation in the next few lessons is between myself and ChatGPT …
Lesson 20 DFQ: What is the most interesting idea, or insight, you personally discerned from this lesson?
Next Lesson: Garbage Prompts
The most interesting idea for me is that, however technology evolves, and whatever the advances in AI, the human role would possibly remain critical. For me personally, the biggest worry is that the same AI tools that humankind develops, may in the end lead to the demise of good judgment, functionality, and, ultimately, our ability to think. Therefore, humans need to stay sharp and alert, and be able to introduce the right inputs so as to generate the right results. However, this takes effort to achieve, and we need to be intelligent about it. I quote this excerpt which I found highly meaningful: ”You can’t stand still if you want to keep up! In an increasingly high-tech world, we may well feel that we must become more intelligent just to hold our own. To make the most of our mental abilities, we need to understand the latest thinking on what intelligence is— and how it can be developed. (Philips, C. ‘’IQ tester: Boost your Intelligence)
The GIGO framing hit home — the quality of AI output really reflects the quality of your thinking. Bad input is still bad, just faster. That’s both a warning and an invitation.
Two insights rather than one for me. Firstly, reading 10 “goods” for something I think is a crap idea made me realise how quickly I shut down my objective viewpoints. Secondly – even though I have used it a fair bit – Chat GPT is way better at reasoning than I had imagined.
Very important is the response from Chat GOT that tells us how best to ask it questions – by providing it with relevant information and context when asking your question and avoiding ambiguous language.
Writing clean and concise questions with context and contraints to minimize hallucinations is critical to receiving high quality responses to a prompt from ChatGPT.
This is both eye opening and refreshing. Data has always been something collected but to what means is another thing? We gain access to many a bias when we collect on topics that reinforce our own map of the world. Chat gpt is, likewise, collecting our data about the world and spitting out what we want to hear. Fascinating.
Do I know how to question well as a result of reflecting on the lesson? I really don’t know. I better explore some more.
I gleamed from this lesson that the enormous power of Large Lanugage Models like ChatGPT, lies in one’s skill in asking questions in a manner that is clear, precise and depending on what channel you are working in, i.e. reseach, fact finding or creative, “imaginative”, to get optimum “needle moving” responses from LLMs.